lead-scoring-model
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security threats detected. The skill follows best practices for handling untrusted data by explicitly instructing the agent to treat external content (like lead lists) as data only and ignore any embedded instructions.
- [COMMAND_EXECUTION]: The skill uses a bundled Python script (
scripts/lead_score.py) to perform complex mathematical scoring. This script is part of the skill package, uses only standard libraries, and does not perform network or file system operations (other than reading the input JSON). - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes lead lists and ICP criteria provided by users. However, it includes robust boundary instructions and sanitization logic (using JSON parsing) to mitigate this risk.
Audit Metadata